Performance Analysis Of Deep Neural Network Algorithm With Optimizer For Rumour Detection

Document Type : Primary Research paper


Department of Computer Science and Technology,Central University of Jharkhand, Ranchi, India


Social networking sites and social media are vital sources of diverse nature of
information. Enormous amount of data is floating per second in cyberspace through the
internet. Numbers of social media applications are used to propagate and maintain the
information. In most of the cases, these applications are also being used to spread false
information and rumours that affect the individual and the society abruptly. In order to
reduce the harmful impacts of rumour an automated rumour detection system is required.
Several efforts are being made and various mechanisms have been developed to find out
authenticity of information and dispel rumors on social networking sites by assessing their
content and social circumstances with machine learning and deep learning approaches. In
this paper, we have performed a comparative analysis of two deep learning models LSTM
and BiLSTM with Adam and RmsProp optimizers to detect and track the rumour or nonrumour
text from the given dataset.